[ros-dev] Re: [ros-svn] [gdalsnes] 18113: -reorder
InsertXscendingOrder macro argument order and update uses
w3seek at reactos.com
Tue Sep 27 02:22:26 CEST 2005
Alex Ionescu wrote:
> I'm going to have to voice my extreme outcry against this patch. It is
> replacing clear and easily debuggable code by non-standard macros which
> are impossible to debug. It is also going to create problems when people
> which are not used to them (such as myself):
> 1) Have no idea what some LIST_FOR_EACH_SAFE thing is doing vs
> LIST_FOR_EACH vs InsertListLifoSafeMaybeLoopOrWhoKnowsWhat
> 2) Will never code using these macros, and thus the source will become
> full of both methods, which is totally ugly.
> Sometimes, especially in kernel code, it's much better to have an
> expanded logic then to use a macro. I wrote a great majority of the code
> that has just been, imho, pollutted with these macros, and I don't
> appreciate for someone to barge in and change my code for some macro
> that I don't understand, nor wish to, because it only causes problems. I
> take great care of the code I write and "own" it (in the programming
> term) so that if a bug in it arises, I can quickly identify it and be
> held responsible for it. The addition of these macros hinders that effort.
> I don't mind someone using them inside his or her own code. I just mind
> their usage in ntoskrnl, and even more so when it overwrites the clean,
> clear code that I've written and am used to debugging.
> Best regards,
> Alex Ionescu
I definitely agree.
More information about the Ros-dev